Automated estimation of voxel size using reference pixel size in phenix?
![](https://secure.gravatar.com/avatar/dc4f7ac3fa1a4c4313492c61dca1b11d.jpg?s=120&d=mm&r=g)
Hi, In cryoEM, the pixel size is often not known with certainty, and we need to calibrate it by measuring the real space correlation of a reference crystal structure of a component at different nominal pixel sizes. Currently we do this manually, using `fitmap` in UCSF Chimera, but given the sophisticated set of tools in Phenix for fitting atomic models and analyzing real space correlations, I was wondering if a tool in Phenix already exists to do this, or whether there would be any interest in adding such a tool at some point? Cheers Oli
![](https://secure.gravatar.com/avatar/361c7059a29c3e70afe24946a6b12abb.jpg?s=120&d=mm&r=g)
Hi Oli, The relatively new EM_placement docking tool (phenix.voyager.em_placement) has an option to refine the voxel size as part of the likelihood-based rigid-body refinement step, and it seems to work pretty well. At the moment, this has to be done as part of a docking search, but I’m planning to implement a rigid-body refinement option for previously-placed models, which might be useful in some circumstances. Best wishes, Randy
On 13 Nov 2023, at 13:42, Oliver Clarke
wrote: Hi,
In cryoEM, the pixel size is often not known with certainty, and we need to calibrate it by measuring the real space correlation of a reference crystal structure of a component at different nominal pixel sizes.
Currently we do this manually, using `fitmap` in UCSF Chimera, but given the sophisticated set of tools in Phenix for fitting atomic models and analyzing real space correlations, I was wondering if a tool in Phenix already exists to do this, or whether there would be any interest in adding such a tool at some point?
Cheers Oli _______________________________________________ phenixbb mailing list [email protected] http://phenix-online.org/mailman/listinfo/phenixbb Unsubscribe: [email protected]
----- Randy J. Read Department of Haematology, University of Cambridge Cambridge Institute for Medical Research Tel: +44 1223 336500 The Keith Peters Building Hills Road E-mail: [email protected] Cambridge CB2 0XY, U.K. www-structmed.cimr.cam.ac.uk
![](https://secure.gravatar.com/avatar/dc4f7ac3fa1a4c4313492c61dca1b11d.jpg?s=120&d=mm&r=g)
Oh wonderful, that is exactly what I was looking for, I will test it now! Thanks Randy! Cheers Oli
On Nov 13, 2023, at 10:30 AM, Randy John Read
wrote: Hi Oli,
The relatively new EM_placement docking tool (phenix.voyager.em_placement) has an option to refine the voxel size as part of the likelihood-based rigid-body refinement step, and it seems to work pretty well. At the moment, this has to be done as part of a docking search, but I’m planning to implement a rigid-body refinement option for previously-placed models, which might be useful in some circumstances.
Best wishes,
Randy
On 13 Nov 2023, at 13:42, Oliver Clarke
wrote: Hi,
In cryoEM, the pixel size is often not known with certainty, and we need to calibrate it by measuring the real space correlation of a reference crystal structure of a component at different nominal pixel sizes.
Currently we do this manually, using `fitmap` in UCSF Chimera, but given the sophisticated set of tools in Phenix for fitting atomic models and analyzing real space correlations, I was wondering if a tool in Phenix already exists to do this, or whether there would be any interest in adding such a tool at some point?
Cheers Oli _______________________________________________ phenixbb mailing list [email protected] http://phenix-online.org/mailman/listinfo/phenixbb Unsubscribe: [email protected]
----- Randy J. Read Department of Haematology, University of Cambridge Cambridge Institute for Medical Research Tel: +44 1223 336500 The Keith Peters Building Hills Road E-mail: [email protected] Cambridge CB2 0XY, U.K. www-structmed.cimr.cam.ac.uk
![](https://secure.gravatar.com/avatar/dc4f7ac3fa1a4c4313492c61dca1b11d.jpg?s=120&d=mm&r=g)
Is that the option "refine_cell_scale"? Or is there another option I am overlooking Cheers Oli
On Nov 13, 2023, at 10:41 AM, Oliver Clarke
wrote: Oh wonderful, that is exactly what I was looking for, I will test it now! Thanks Randy!
Cheers Oli
On Nov 13, 2023, at 10:30 AM, Randy John Read
wrote: Hi Oli,
The relatively new EM_placement docking tool (phenix.voyager.em_placement) has an option to refine the voxel size as part of the likelihood-based rigid-body refinement step, and it seems to work pretty well. At the moment, this has to be done as part of a docking search, but I’m planning to implement a rigid-body refinement option for previously-placed models, which might be useful in some circumstances.
Best wishes,
Randy
On 13 Nov 2023, at 13:42, Oliver Clarke
wrote: Hi,
In cryoEM, the pixel size is often not known with certainty, and we need to calibrate it by measuring the real space correlation of a reference crystal structure of a component at different nominal pixel sizes.
Currently we do this manually, using `fitmap` in UCSF Chimera, but given the sophisticated set of tools in Phenix for fitting atomic models and analyzing real space correlations, I was wondering if a tool in Phenix already exists to do this, or whether there would be any interest in adding such a tool at some point?
Cheers Oli _______________________________________________ phenixbb mailing list [email protected] http://phenix-online.org/mailman/listinfo/phenixbb Unsubscribe: [email protected]
----- Randy J. Read Department of Haematology, University of Cambridge Cambridge Institute for Medical Research Tel: +44 1223 336500 The Keith Peters Building Hills Road E-mail: [email protected] Cambridge CB2 0XY, U.K. www-structmed.cimr.cam.ac.uk
![](https://secure.gravatar.com/avatar/361c7059a29c3e70afe24946a6b12abb.jpg?s=120&d=mm&r=g)
Yes, that’s the right option! It doesn’t allow you to specify a starting scale value, which defaults to 1. That might be a useful option for cases where the initial docking fails because the scale is too far from the correct value. My impression is that this used to be a bigger problem in the past, but that microscopes are generally calibrated reasonably well these days. Best wishes, Randy
On 13 Nov 2023, at 15:45, Oliver Clarke
wrote: Is that the option "refine_cell_scale"? Or is there another option I am overlooking
Cheers Oli
On Nov 13, 2023, at 10:41 AM, Oliver Clarke
wrote: Oh wonderful, that is exactly what I was looking for, I will test it now! Thanks Randy!
Cheers Oli
On Nov 13, 2023, at 10:30 AM, Randy John Read
wrote: Hi Oli,
The relatively new EM_placement docking tool (phenix.voyager.em_placement) has an option to refine the voxel size as part of the likelihood-based rigid-body refinement step, and it seems to work pretty well. At the moment, this has to be done as part of a docking search, but I’m planning to implement a rigid-body refinement option for previously-placed models, which might be useful in some circumstances.
Best wishes,
Randy
On 13 Nov 2023, at 13:42, Oliver Clarke
wrote: Hi,
In cryoEM, the pixel size is often not known with certainty, and we need to calibrate it by measuring the real space correlation of a reference crystal structure of a component at different nominal pixel sizes.
Currently we do this manually, using `fitmap` in UCSF Chimera, but given the sophisticated set of tools in Phenix for fitting atomic models and analyzing real space correlations, I was wondering if a tool in Phenix already exists to do this, or whether there would be any interest in adding such a tool at some point?
Cheers Oli _______________________________________________ phenixbb mailing list [email protected] http://phenix-online.org/mailman/listinfo/phenixbb Unsubscribe: [email protected]
----- Randy J. Read Department of Haematology, University of Cambridge Cambridge Institute for Medical Research Tel: +44 1223 336500 The Keith Peters Building Hills Road E-mail: [email protected] Cambridge CB2 0XY, U.K. www-structmed.cimr.cam.ac.uk
----- Randy J. Read Department of Haematology, University of Cambridge Cambridge Institute for Medical Research Tel: +44 1223 336500 The Keith Peters Building Hills Road E-mail: [email protected] Cambridge CB2 0XY, U.K. www-structmed.cimr.cam.ac.uk
participants (2)
-
Oliver Clarke
-
Randy John Read